GitGyun/visual_token_matching
[ICLR'23 Oral] Universal Few-shot Learning of Dense Prediction Tasks with Visual Token Matching
This project helps researchers and engineers quickly train computer vision models for complex tasks like depth estimation or semantic segmentation, even with very little data. You provide a small set of labeled images, and the system outputs a trained model capable of performing detailed pixel-level predictions on new, unseen images. It's ideal for those working in areas like robotics, autonomous vehicles, or medical imaging who need to adapt models to specific environments or datasets efficiently.
255 stars. No commits in the last 6 months.
Use this if you need to develop accurate image analysis models for tasks like identifying objects, edges, or depth, but have limited labeled data for training.
Not ideal if your main goal is simple image classification (e.g., categorizing an entire image) rather than detailed pixel-level predictions, or if you have abundant labeled data.
Stars
255
Forks
13
Language
Python
License
MIT
Category
Last pushed
Oct 13, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/GitGyun/visual_token_matching"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
jakesnell/prototypical-networks
Code for the NeurIPS 2017 Paper "Prototypical Networks for Few-shot Learning"
harveyslash/Facial-Similarity-with-Siamese-Networks-in-Pytorch
Implementing Siamese networks with a contrastive loss for similarity learning
oscarknagg/few-shot
Repository for few-shot learning machine learning projects
google-research/meta-dataset
A dataset of datasets for learning to learn from few examples
Sha-Lab/FEAT
The code repository for "Few-Shot Learning via Embedding Adaptation with Set-to-Set Functions"